Consequently, many States began to raise the drinking age and by 1986, 40 States again had a minimum drinking age of 21. While there was considerable variation among States in studies of the impact of this change, most States showed that increasing the drinking age was associated with fewer traffic accidents among 18- to 20-year-old drivers. Studies conducted between 1972 and 1984, while varying in the population (male versus all drivers) and type of accident, found that the higher drinking age was associated with reductions of 7 to 28 percent in accidents, particularly fatal crashes, among 18- to 20-year-olds. While most studies examined the effects on male drivers, some studies suggest that legislative changes may have had even greater effect on reducing crashes of female drivers. Some evidence also has been found that changes may reduce accident involvement among drivers younger than those directly affected by the higher drinking age. Finally, some studies suggest that raising the drinking age in one State may lead to increased crashes in border areas of neighboring States with lower alcohol purchase ages. However, such increases appear to be overwhelmed by Statewide reductions. 2 tables and 36 references.